An ML model predicts customer churn well on training data but fails on new data. Which change is most likely to fix this overfitting problem?

Overfitting occurs when a model learns the training data too well, including noise, leading to poor performance on unseen data. Increasing the regularization parameter penalizes large coefficients, forcing the model to be simpler and less sensitive to minor fluctuations in the training data. This reduces its complexity and improves its ability to generalize to new data. Decreasing regularization would allow the model to become even more complex, exacerbating overfitting. Adding more input features without careful selection can increase complexity and potentially worsen overfitting. Training for more epochs can also lead to overfitting if the model has already learned the underlying patterns and is now memorizing noise.